Joseph Makin, Assistant Professor of Electrical and Computer Engineering

Joseph Makin, assistant professor in Purdue University’s Elmore Family School of Electrical and Computer Engineering, has received a National Science Foundation (NSF) CAREER Award. These prestigious awards support junior faculty who exemplify the role of teacher-scholars through research and education, and the integration of these endeavors in the context of their organizations' missions. The awards, presented once each year, include a federal grant for research and education activities for five consecutive years.

The grant will support Makin’s project, entitled “Brain-Machine Interfaces for Speech.” This project is focused on helping people who have lost the ability to speak, such as those affected by ALS, strokes, or brain injuries, by using electrical signals from their brain to understand what they want to say. While some early systems have been tested in research, they’re not yet accurate enough to replace simpler, non-invasive methods like eye-tracking or typing with assistive devices. The goal of this project is to significantly improve these brain-based speech decoders by gathering more data and using it more effectively with advanced machine learning techniques.

One challenge is that data from inside the brain (iEEG) is hard to collect. To overcome this, the team is using methods that train the system on tasks related to brain signals, even when no speech is involved. They’re also using speech and language models (similar to those used in voice assistants) to help the system learn to predict speech from brain signals. Another strategy is combining data from many participants to train a more powerful model. By working with volunteers from multiple hospitals and research centers, they hope to gather enough data to make these systems more accurate and reliable.

"This line of research has strong potential to improve the lives of people in the near term, by bringing cutting-edge machine learning to bear on neuroscience,” said Makin. “My research group is excited to work in this area and we're very glad that NSF shares our enthusiasm."

In addition to the research, the project will also include educational efforts to train a new generation of scientists in both neuroscience and machine learning, making the technology behind brain-machine interfaces more accessible to a broader audience.